In this episode of the Consumer Insights Podcast, Thor is joined by Steve Mushkin, Founder of Latitude.

What could you do with live, actionable insights directly from your audience at the very moment they see your content?
Technology today makes it possible to gather, analyze and implement viewer feedback in real time, and these methodologies have the potential to bring greater depth and granularity to audience insights.
We spoke with Steve Mushkin, Founder of Latitude about his decades of experience in audience and content research to learn more about how brands can use cutting-edge technology to deepen customer understanding and increase business impact.Join us as we discuss:

• The difference between tactical and fundamental insights and how to leverage each
• Developing a high-level vision to guide team and individual decisions and innovation
• Building a powerful experience for your audience as the attention economy evolves
• The evolution of content research — and what it means for brands today

You can access all episodes of the Consumer Insights Podcast on Apple, Spotify, or Spreaker. Below, you'll find a lightly edited transcript of this episode.

Introducing Steve

Thor Olof Philogène:

Hello everyone and welcome to The Consumer Insights Podcast. Today, I'm excited to have a brilliant insights leader joining me for what I know will be an illuminating conversation. I'm thrilled to introduce today's guest, Steve Mushkin, founder of Latitude, a leading insights provider for digital content and technology. Through its interactive video suite, Lumiere Latitude helps the world's foremost media and technology companies understand how the evolving online and mobile landscape, as well as emerging user behaviors, translate into new business opportunities. With over 35 years in content research space, Steve started his career leading program and audience research at HBO and Comedy Central. He also has a master's degree in economic international development from the University of Geneva, as well as a bachelor's degree, European History in Comparative Literature and Film from Stanford University. Thank you so much for joining me, Steve.

Steve Mushkin:

It's an honor to be here. Thanks for inviting me.

Thor:

Now, a lot of us are really excited to get to know you. So could you take a couple of minutes to tell us about yourself, your company, Latitude, and how you got to where you are today? How did it all begin?

Steve:

I will start with HBO. I began my career. I won't start earlier than that. That's probably professional life is a good starting point. Although there are some interesting things before that. I began my career with HBO and Cinemax and nicely now they've combined into the new name Max. I had a little bit of a melancholy when the name HBO disappeared a month or two ago. I was working for HBO and Cinemax at the time, working in the on-air promotion department and doing basically a comprehensive database of all the on-air promos and trailers and the detailed breakdown of what they consisted of and how they related to the HBO and Cinemax programming at the time.

After about a year of that, I realized that I was very interested in research and found a very interesting job in the research department at HBO doing work on Original Programming, which was nascent at the time. It was maybe the mid-90s and HBO had really just sort of gotten at sea legs with their original programming and was starting the original series and the movies.

So I got deeply into that and the audience psychology of that. And it was the first golden age, if you will, of HBO's growth in originals. HBO had a golden age prior to that and stayed there for five years, moved on to Comedy Central as head of all programming research and was really able at that point to help Comedy Central in a very happy way. Comedy Central was a happy place to be. There were about a hundred employees. It felt like what young tech companies feel like now, but in the mid-90s. And we just had a lot of fun and obviously the name of the network and the content of the network helped with that. And so Comedy Central at that point was evolving from a baby TV network that was the combination of Time Warner and Viacom to a network that was more sophisticated with long-form programming like South Park. So I was able to work with the management team to really evolve Comedy Central into the company slash network that it is today. Obviously since then they've made quite a bit of progress. And then in 1996 I founded Latitude and that was basically when the internet became mainstream. And it was clear to me at the time that there was going to be a real influence of content and technology, which has continued to this day. And so we centered the company around that essentially both in terms of technology experiences, as well as content experiences, of course, digital content experiences, which were really evolving and just beginning many of them. And so there was a pretty wide open space to understand how people were experiencing the web and content on the web as well as the technological underpinnings of that. And since then the company's grown and really deepened its understanding and deepened our work with many clients around that intersection of content and technology. And by content, I'm speaking broadly of both pure program content, digital content at various times, and advertising and branded content. So that's a semi-brief overview.

Perspectives from a career in innovation and strategy

Thor:

And that's quite a career you've had so far and you know, fascinating your companies you've worked out and with. And given where you sit in the world of insights, how would you define an insight?

Steve:

I would say that there are many different kinds of insight, at least in our experience with our work and the clients we work with. And some of the insights are, and I'll divide them into multiple categories, if you will. The first is probably the most obvious, which is something that one desired to learn that was in the wheelhouse of the study or the system that's being studied and on which one would take action. So it's a very tactical kind of insight and one that was being searched for from the get-go, so to speak. Second kind of insight is foundational insights, I would use the term. And that kind of insight is sort of underlying principles that may be discovered or maybe we searched for them that affect multiple behaviors, but they're strategic or even sometimes beyond strategic in terms of affecting the entire experience of the company and their offerings. And so those kind of larger insights can happen serendipitously or they can happen in very explicit ways. And I'll just give you one example, which happened probably about 15 years ago. We were working with Court TV, now truTV, and they were one of our most treasured and appealing clients in a variety of ways. We had a great relationship with them and the people making the programming there. And we did a segmentation study and we were sort of understanding the relationship between viewers and the type of content that Court TV, which was both courtroom programming at the time, as well as fictional and non-fictional shows in the evening.

And we came across an insight which served them well for quite a while, which was that there are basically three types of viewers for their type of programming. Viewers that identified with the cops or investigators, detectives, the sort of police broadly writ. Viewers who identified with the criminals and or the wrongdoers, if you wanna use a slightly softer word. And viewers who identified with the victims or the average person.

So there were sort of three roles that we understood through the segmentation and we were looking for it, but we didn't know that we would find it that clearly. And so we designed the work around exploring roles, but it wasn't obvious that this clarity was going to happen nor that the power of the identification with the criminal was going to be so high and so strong. And so Court TV slash True, because they were in transition at that moment, really sort of developed some of their programming and strategic directions around that segmentation based insight. And it was great because it's incredibly intuitive, right? I mean, when you think about it, there are three major roles. And of course there's crossover among those roles. So you get some interesting places with ops that might be corrupt or justice system that is not working perfectly or criminals that might not be as criminal at first glance one would imagine and victims that have agency and power in some cases. So the crossover actually got to really interesting places and we worked with the network to leverage that in their characters, in their programming and so on. So that was a good example of a larger insight that was a fundamental insight and not a tactical one although certainly it ended up being used tactically.

How they drive strategic business decisions

Thor:

That's a fascinating example. And I think to your point, I mean, it does make sense, but it must have been incredible to uncover. And building on that, from your perspective, you've highlighted it on the strategic one, the foundational insights. But why do you think that marketing consumer insights are so important? What is it that on a general level, they allow businesses to do?

Steve:

Obviously, there's the straightforward aspect of decision making. And I just mentioned in a tactical and or strategic sense, I think the more powerful thing on occasion, and it doesn't always happen, but when it does, it's beautiful to speak is that you have the opportunity to create vision for a business and that vision is more powerful than a tactical decision. And I'll get to an example or two in a moment,

but there's vision that's at a higher level sometimes than individual decisions. And obviously the decisions flow from that kind of vision. And it can lead to innovation that can be really important for any company that's working in the arena that we're in, because it's changing all the time.

And of course, at this point, any company needs to be innovating pretty consistently. So the idea of having insights lead to vision is very strong. And just to give you a sort of fairly recent example, and this is sort of working synchronously, if you will, NBC Universal, with whom we work quite consistently on advertising innovation and there's individual decisions being made and individual strategic considerations. But the larger picture is that NBCU has a vision of the relationship of brands, advertisers and content that is very cutting edge, very sophisticated and allows them to innovate in some new and powerful ways. Just to give you an example, because the vision then plays into very specific manifestations in the content and the effects on brands. So NBCU has across their properties and their content arenas, they've got 10 plus advertising innovations that they're implementing at this point and have been for the past few years, both in streaming and in linear and doing things like something I was just reading about last night, actually, where they try to create a more seamless relationship between the advertising pod and the content itself. And, you know, advertising being thought about in seamless way as content in many cases, they're doing something that's very interesting where they'll do a stop screen or pause on the last image of a scene. They'll hold that image. They'll dissolve into an ad that's fairly seamless with that image and then come right back out of the ad into the stop screen image again and continue the episode. And obviously the length of that occurrence can affect the viewer experience significantly. So if it's five seconds, it's really seamless. If it's 30, a little different. But they're really exploring with that and other examples. They saw another one recently where they use a show within a show kind of thing. It's almost meta of an advertisement that's inside the show. That's not just a product placement, but it's more sophisticated insertion of another sort of little show within, like when Somebody's reading a book in a movie kind of thing. Right. And you see the name of the book. So and it's obviously done for a different purpose. But really interesting that they're thinking in fairly, you know, some of this in many would straightforwardly after the fact, seems clear. And why didn't Somebody think of that before on the one hand? And on the other hand, it's not. Somebody needs to think about it and NBC's vision and their constant desire for innovation on the ad side and on the ad experience and how that helps brands and how that allows brands to sort of connect with the content in elegant ways is pretty powerful, right, because it creates across NBC's entire offering across their properties, across their different platforms, an experience for viewers that's ultimately more integrated and more powerful. You know, insights for at that larger level in terms of creating a vision for something that's feature oriented is can be really valuable for a company like that.

Unlocking competitive advantage: the technology behind Latitude and Lumiere's insight solutions

Thor:

I think the work you're doing at Latitude and Lumiere is fascinating and I think incredibly important for brands as the attention economy continues to evolve and become more competitive, especially with the rise of streaming platforms and the always on world that we live in. Could you tell us a bit more about how your technology helps companies capture and leverage insights?

Steve:

So regarding NBC, just to take a good example since I was just speaking about them, in many of the studies we do with them and the ongoing projects, we use Lumiere, which I'll explain in depth in a moment if it feels appropriate. We gather real-time feedback from viewers about the content. And so it's not real-time on NBC, you or NBC properties, just to be clear, but it's happening in real-time as the viewer watches in our environment. So just to take the recent example I mentioned just a moment ago, we show the various innovative ad experiences and branding experiences on NBC use properties and take whether it's a five-minute clip or a longer and sometimes shorter. We'll show that in Lumiere and let people react either using answering questions, leaving emojis right on the content at a moment that they like, leaving comments in an open-ended way, so a combination of quantitative and qualitative that they can provide to give feedback to NBCU, the advertisers, the brands, and us, of course, as the middlemen in terms of how they're experiencing that innovative commercial pod, how they're experiencing the brand at that moment in time. So there's a moment-to-moment aspect to it that's important. And then there's a granularity because we can capture that information at the exact moment that it's happening, both in terms of temporal aspect, the moment that it happens.

So when the transition goes from that, you know, stopped image that I was talking about a moment ago with the program and into the advertisement, then we can get right there at the moment where there's that pause and the transition.

We can get the viewer reaction. So it's not later. It's not 10 minutes after the fact. It's not 10 days after the fact. It's right then that we get the viewer reaction in detail. And the viewers can click at any point on the screen so they can react not only temporally, but spatially to anything they're watching.

Fueling breakthroughs: how Lumiere’s insights unlocked new opportunities for businesses

Thor:

Let's dig a bit deeper there. I mean, do you have any stories you can share to illustrate how insights sourced from Lumieres have fueled innovation and help us understand what the insight was that created the opportunity, how you identified it and take us all the way to the end result.

Steve:

So obviously with NBCU, it's an ongoing process to continue to innovate with advertising and support the innovation, right? They're going out on a limb in some cases, doing some new things that may or may not work. And we're diagnosing as well as showing the value of that kind of innovation. Another example is something we did and it's very different, but semi-related or contiguous, is recently did some work for ESPN on a series of content series they had called 50-50. It was about the 50th Anniversary of Title IX, which is the women's equality in sports. So Title IX stipulated 50 years ago that women that no state-funded institution could have any sort of inequality or discrimination, not reading exactly from the language, but inequality was basically designed to abolish inequality or discrimination against women in sports. And that has had immense success in terms of allowing women's sports to flourish in ways that were just not happening and not imaginable even 50 years ago. And we've seen the positive consequences of that. So as part of 50-50, ESPN asked us to help understand the positive effects of the content and the positive effects of the content on the brands that were supporting it. One of those brands was Champion, which sports clothing and apparel manufacturer provider, and Champion was heavily involved. There were other sponsors as well, but we did, we focused on Champion and I won't go into all the details and metrics. But basically there was a virtuous circle and of course, especially women, but not only, really diving deep into the content and what ESPN was providing and having very positive brand effects on not only ESPN, which is of course a direct effect from the content, but from with Champion itself. So Champion had a double positive effect. It had the effect of being in the environment and having all the cache of the positive effects of 50-50, as well as the direct brand effectiveness and ad effectiveness positive effects that one looks for in these kinds of things generally.

So there were direct effects in terms of interest in the brand, but there were also called them indirect effects in terms of brand attributes and the overall brand halo as a result of being associated with 50-50. So the positive sort of influence on women's sports became part of Champion's brand for that period and hopefully for a longer period. All right, no one would imagine that remains.

And so Champion in particular, because we took a close look there, was really positively affected. We gathered that through Lumiere, just to circle back to the direct answer to your question. And we did it in a really nice way. We had a combination of a controlled experiment and the direct feedback that I was talking about earlier, where we had some people who weren't exposed to any of the content and we compared them to people who were exposed and gathered reactions in Lumiere through that exposure, both in terms of the content itself of the 50-50 programs about different women and their success stories and different sports and the successes that have happened there in different universities and other organizations where the positive effects of 50-50 played out, as well as, of course, the direct and indirect positive effects on Champion that we captured in Lumiere in the moment of watching Champion's sponsorships of 50-50. So it was a real success story on all levels and one that on the one hand one might expect, but the positive effects were even stronger than we would have hoped.

35 years of change: how content research has evolved from HBO to the streaming era

Thor:

been in the world of audience and content research for quite some time. You started out at HBO and Comedy Central before founding Latitude. And in over 35 years that you've spent in this space, how do you think content research has evolved? I mean, for example, in series, movies and branded content.

Steve:

So it's evolved in a way that's occasionally been difficult for us in that there's a great deal of data coming in from the wild that didn't exist when I started the company and started off my career. So back when I first began at HBO, you had variations of Nielsen ratings and a few other things that people were attempting to start at the time. Now we have tremendous amounts of data pouring in every moment and not in periodic ways as it was at the time. They've got sample sizes. When I started my career, Nielsen's family sample was 5,000 households across the U.S. And I'll tell you a funny anecdote related to that in a moment. So the sample sizes have gone from 5,000 to, depending on the streaming service or the network, 10 plus million. And so Netflix is the amount of data that Netflix has coming in every day dwarfs the amount of data that the entire television universe had in 1992. So that is a tremendous competition for us in terms of primary market research. And what we do as a result is attempt to combine our insights and our data with other forms of data so that we can obviously leverage the advantages of that. So the big change, at least certainly the one that comes to mind most immediately, is that obvious one, which is the data is coming through fast and furious in ways that are proprietary to each content provider. And YouTube is another variation of that, the world's content and video content gorilla, because we're focused on video content. YouTube is a fantastic source of information for all its creators, sort of aggregate sense, as well as data for the individual creators.

And so for us, the other, I think the second answer to that question that's really interesting sort of paired with the data flow is that we have seen a change in how content research occurs. It is no longer a sort of longer chronological sequence of do a pilot test, decide whether to make a show, film a couple of episodes, make some more, do some more research, make some more decisions. It's become much more of a streaming type of, if you will, a streaming flow of research. And so everything happens more quickly. And the cycle of production to data back into production is often immediate.

Certainly you see that in spades in YouTube, right? But you also see it with big producers who are constantly adjusting what they're doing, not in real time on this large Hollywood movie, but certainly much faster than they used to. And you certainly see this with the advent of data flowing in from social media as well. So, you know, the biggest Hollywood producers are adjusting what they're doing with their marketing and trailers based on what's going on in social media. That speed of the data flow and the data cycle back into the content production has changed immensely over the 20 plus years that I've been doing this and that the company's been doing this. It's interesting. It makes things really sort of fluid and dynamic, but it also occasionally, on some days, makes things more difficult.

AI’s role in shaping business insights: opportunities for growth and key pitfalls to avoid

Thor:

And on the topic of change, AI is obviously having a big moment right now. And it's something that's on the minds of a lot of insights leaders. As the founder of a company that delivers AI generated insights, where do you see the opportunities for AI insights working? Conversely, where do you see the potential pitfalls?

Steve:

So I'm an AI optimist and fortunately as a growing camp of those. And so the opportunities, we haven't even begun to scratch the surface, but just to outline a couple, it's very clear at this point that AI analytics are beyond even our wildest dreams. And we are able, for instance, to use ChatGPT. We're using 3.5 right now because it's actually a little more stable than 4.0. We use it inside Lumiere to analyze open-ended content comments. So the open-ended comments we analyzed with some pretty sophisticated NLP, which is AI-based Natural Language Processing and AI-based. We were doing for the past five or so years. It's gone up a level in no time flat. We're able to provide X, you know, English or other languages, multiple languages, which is also an advanced. We're able to provide summaries of insights from open-end from tremendous amounts of qualitative data, synthesized seamlessly in an automated way. That would have been unthinkable five years ago. I remember going to a SMR conference. I think it was in Berlin where people were talking about NLP and everyone was getting excited about it. It turned out that it was pretty hard to do at that time, but now it's not.

The analysis capabilities are really a step ahead or five steps ahead of what they were. The second thing that's really a big opportunity with AI is the generative possibilities. And that's the name Generative AI is really appropriate. So the ability to generate, for instance, new ideas and new concepts, both to execute on as well as to test is pretty sophisticated at this point. And so we're using AI at the moment to do that sort of thing so that we can generate at scale possibilities that can be researched and can develop insights on things that haven't been done, but are examples of what might be done.

And so we're able to generate, you know, some people use the term synthetic data. I would call it in our case, you might call it synthetic content. And synthetic sometimes has a negative connotation, but in this case, positive one where we're able to generate possibilities and use those as part of the research. And that gets really interesting. Mark Anderson, who's the former writer of Netscape and now the head of Anderson Horowitz, the VC firm, has talked about that in some really interesting ways where the next level of AI is AI generating in ways that create a feedback loop back to humans. And so it's not just humans conversing with AI, but AI generating in ways. And you see this with all sorts of things in the content space, not only generated text, but also generated video and images, of course. So we're using that as new ways of looking at research itself and generating insights from Generative AI, which is pretty difficult to think about when I was starting my career, right? And I got excited about Generative AI 10 years ago when I first heard about it. We just didn't know when it was going to come. So we're using it in that way directly in our work and in Lumiere and currently looking at new ways to play with that. In terms of some of the potential negatives and pitfalls, just to answer the second part of your question, I think one of them is not so worried, I should preface, about things like AI taking over the world, that kind of large-scale thing. But just in the research world, I think what is probably going to be important is you really understand how to think sharply about this and use it in ways that leverage its capabilities. So Generative AI is great at generating. And so the example I just gave you is something that I think many research firms and brands and advertisers can use in all sorts of ways and are beginning to. I think where the major pitfall might be is allowing the AI to do everything by itself without some co-piloting. I have a Tesla and I let it drive itself some of the time with my oversight. And some of the time I just decide to take control of myself completely. And I moved back and forth in modes between sort of co-piloting, partial co-piloting and so on. And I think that's a good, interesting model for AI going forward in research and insights is that. Well, you know, the humans will continue to be able to do some really high-level thing, have vision, as I referred to earlier, and let AI do some things that we formerly did, but AI can do as well or better. And I think the main pitfall is just to sort of not that AI is dangerous and certainly in the research space, I think that's less of an issue, but simply that we think about what are the most powerful ways to leverage it and act intelligently on those. Where's the highest and best use of AI? I guess it's my simple way of saying it and being thoughtful about that. And it's evolving. So we're going to constantly be looking at new opportunities.

The ultimate lunch: a leader’s dream guest in the world of insights

Thor:

Thank you so much for that. See, it really hurts me, but we've come to the end of this episode. And there's only one question that I want to ask you before that, which is a question that is a favorite of mine, which is who in the world of insights, would you love to have lunch with?

Steve:

So I'm going to answer that question in an unusual way. The person that I have seen, two people I've seen that I think are actually not directly

in the world of insight, but certainly have great insights and are operating within our realm, are a guy named Ben Thompson, who writes the blogs for Techory. And he is incredibly insightful about the intersection of content and technology. And though he's not an insights professional in the traditional way, so to speak, but phenomenal business thinker and has great vision into where content and technology and their intersection are going.

So he's the first of my answers. The second one would be Satya Nadella, who's the head of Microsoft, who has taken that company to places that I think seven or 10 years ago, people weren't really expecting, and beautiful job of it. And I think he's done it in great part because he has a really insightful feel for what his users, audiences, and potential audiences across the range is Microsoft products. And Microsoft obviously was not one of the world's most beloved companies 10 or 15 years ago. They're one of our clients and they have, their brand has obviously changed and evolved in some really positive, powerful ways in the past decade under his leadership. And I think in part, not only because he's got a fantastic technical mind, I think in part because he's got a great deal of empathy and a great deal of an ability to combine vision and empathy and empathy with the audiences, users, and whatever else we want to call them, that Microsoft and other technologies are in contact with these days. And he's got empathy in that sense, he's got empathy in a broad sense for the potential of technology and related content and what it can do. He's got a deep understanding of AI and he's got a great feel for how to run an organization. So I admire him at least what I've seen in terms of some of his speaking and some of his writing. And I think that combination in a leader, but also in an insights professional to your question in more direct way, that combination of empathy and vision and expertise is very powerful and spending a good lunch with him would be a nice experience. By the way, I'll just add one more thing if you don't mind. I did have lunch with Somebody who's also not an insights professional, but was on my list, Tim Berners-Lee, the creator of the web. And I had been part of my education in graduate school in Geneva, and he invented the web, the early web as we know it at Geneva at CERN. And so having lunch with him at MIT a few years ago was a great pleasure. We had a good time talking about his vision of the future of the web. So those were insights in a very different kind, but excellent ones. So those are my usual answers to your last question.

Thor:

I think a lot of us were a bit jealous about you having had that lunch, but thank you so much for sharing. Wow, this has been such an incredible conversation, Steve, you have a truly unique perspective on the industry. And I think we all can learn from it. If I rewind and play back some of the moments of our conversation that really stuck with me, I wanted to start off with your definition of an insight, which is you highlighted there are really many different kinds of insights. The first type of insights relates to something that one desires to learn in the wheelhouse of the system being studied on which one would take action. But the second kind of insight is the foundational insight. This one relates to the underlying principles that affect multiple behaviors and are highly strategic in nature. Foundational insights can be immensely powerful as they can allow you to create vision for the business and they can enable innovation. I think we all felt your view of the future as an AI optimist who have a strong belief that we haven't even begun to scratch the surface of the possible. In terms of AI analytics, we're seeing analysis abilities several steps ahead what they were just a few years ago. The ability to generate new ideas and concepts to test, to execute on, is a pretty sophisticated already and can allow us to generate insights on things that haven't been done, but that might be done. And you reminded us that these between brackets Synthetic Insights really open doors for what will be possible next. Now I know I've learned a lot from talking to you today, and I'm sure our audience has as well. So thank you so much for joining.

Steve:

Thank you for having me. It was a pleasure.